A novel ANN-based distribution network state estimator. (May 2019)

• Record Type: Journal Article
• Authors: Ahmad, Fiaz; Tariq, Muhammad; Farooq, Ajmal
• Is Part Of: International journal of electrical power & energy systems. Volume 107(2019)
• Page Start: 200
• Page End: 212
• Publication Date: 2019-05
• Subjects: Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2018.11.019
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: Proposes a novel ANN based state estimator for distribution networks. The algorithm is capable to self train and generates state estimates. The performance of proposed estimator is validated for various network conditions. Tested on two standard IEEE distribution networks and a real life distribution network. Abstract: State estimation is fundamental to energy management systems (EMS) employed in control centers for transmission system networks. However, the automation of power grid requires extension of state estimation to distribution networks, which are yet not observable through the available set of measurement units. In this paper, a novel state estimator based on artificial neural network (ANN) is proposed for obtaining the state of the distribution network despite of the low measurement availability. The proposed estimator was tested for IEEE test distribution networks (33-bus and 69-bus), and for a practical 48-bus Danish network under diverse network conditions such as normal system operation, distributed-generator (DG) connection/disconnection (topology variation), load variation and bad data. Results obtained are promising, which shows the viability of the developed estimator for practical distribution networks.